Excellent tutorial, @terrybritton! Just one question to make sure I understood it correctly:
So the idea is to add extra audio channels (that you keep unassigned) in the audio settings and then add an extra output route to one of those channels to each VST you want to include in the recording. Then you select these output ports under Recording, so that the audio from the relevant VST’s will be recorded into the wav/w64 file?
Yes, select the desired ones from the list at the bottom left that you wish to be included in the recording. If you use a few different setups, though, it can be easy to forget to re-select any you had de-selected!
What I did to get around my “human error” issues was to have certain “standardized” shows I used as templates by making them be unique cantabile instances, and those would have all the recording ports selected in the lower left already so I did not have to ever go in there again. See this post to understand how that is done.
https://www.cantabilesoftware.com/guides/multiConfig
What I do is I create a multi-config and open it so it makes default prefs, then close it and replace its prefs with a basic set that has all my ports already. I then open it and customize it to my new requirements.
If that isn’t clear - open the version you want to get the prefs from, go to Tools=>Open Preferences Folder (or just go to C:\Users\yourprofile\AppData\Local\Topten Software), copy settings.json (and maybe plugins.json + plugins.user.json) and paste those into your new configuration’s folder replacing the others. The rest is easy-peasy! Open it up, load a song or create on, and have fun!
Using that method all the ports you’d created before are already typed in.
Terry
I concur with Terry. I’ve made many recordings with C3, very stable, even with my meager laptop.
Thanks again, Terry. This looks like a very useful way to handle this. I will record tonight’s band practice using all the tips and tricks from this thread (a great thank you to everyone). Tonight, we 'll also be hooking up the drums, and I’m looking forward to playing with the results. Then, if all goes well, our first self-recorded show on Saturday 
I have no problem with the recording process - I’ve used the facility before. But it was the recording across song changes I didn’t know about. As soon as the XR18 arrives I will give it a try. The less gear I have to set up on stage the better - and often we have very little time between set up and performance, so it HAS to be quick and simple.
I recorded a couple of songs last Saturday, and everything went smoothly. I record directly onto my Cantabile laptop and never noticed anything. I did record song for song (which requires me tapping the record button on my controller before and after each song) just to make sure file size would not become a problem.
Another useful trick I found was that you can use binding to mute and unmute microphones and effects from the XR18. I use two different microphones: one when I’m playing my main keyboards and another when playing the keytar, so it’s useful to be able to mute any unused mic without even thinking about it (Song… On Load… etc.).
Running mics through Cantabile incurs a little latency though - if you have a separate midi output you could in theory do this via midi on the X®18 directly.
I’m also pretty sure that the X®18 driver is multi client. At least the X32 one is, I use Cantabile for playback and record in Sonar using the same driver. Just my personal preference to record in a DAW.
No need to run mic through cantabile at all - just control it from Cantabile. That’s how I do it. As soon as backing begins the stage mic is cut from the PA and monitor feeds.
You could maybe run a DAW as well as Cantabile, but… to take the risk of two audio software packages running on one machine - live? I know I wouldn’t! Record in Cantabile, then import to DAW afterwards - worked for me!
Ah I misunderstood then.
I just found it easier to run the DAW in the background and it was very stable, to each their own!